Toward Adaptive Grazing: Leveraging Different Rangeland Monitoring And Assessing Techniques
Biquan Zhao, Post-doctoral research associate, Department of Animal Science, University of Nebraska-Lincoln
Author
05/29/2024
Added
8
Plays
Description
Adaptive Grazing has been aware of a strategy for rangeland management. Implementing adaptive grazing is challenging due to the dynamics in forage production, livestock behaviors, and external environment such as climate change. This talk covers my research in long-term forage production monitoring for plant functional groups, forage biomass estimation by using drone remote sensing, and cattle grazing behavior modeling through GPS collars. Those rangeland monitoring techniques were utilized to enhance our understanding of rangeland dynamics and can help better application of adaptive grazing strategies.
Searchable Transcript
Toggle between list and paragraph view.
- [00:00:00.750]The following presentation is part
- [00:00:02.670]of the Agronomy and Horticulture seminar series
- [00:00:05.441]at the University of Nebraska Lincoln.
- [00:00:08.250]All right, good afternoon everyone.
- [00:00:09.780]Thank you so much for joining us.
- [00:00:12.390]Our semester is pretty well underway, so it is great
- [00:00:15.690]to see such a wonderful turnout here today
- [00:00:18.060]for our seminar.
- [00:00:19.920]I just wanted to do a few orders of business.
- [00:00:23.348]First of all, just wanted to remind everyone
- [00:00:26.040]that our seminars are posted online after the fact.
- [00:00:28.920]So if there's anyone who couldn't make it today,
- [00:00:30.900]please share this with them as well.
- [00:00:33.960]And then in terms of
- [00:00:36.330]how things will progress today,
- [00:00:38.606]Dr. Zhao will present for us
- [00:00:41.220]and then we'll leave questions until the end.
- [00:00:43.830]Because we do have the online audience,
- [00:00:46.290]please wait for the microphone to come to you
- [00:00:48.390]before you ask any questions.
- [00:00:50.430]I'll bring the microphone, you can ask your questions,
- [00:00:52.869]and then we will handle question and answer
- [00:00:56.220]that way at the end.
- [00:00:57.780]All right, so without any further ado,
- [00:01:00.900]I am very happy today
- [00:01:02.670]to introduce Dr. Biquan Zhao.
- [00:01:06.390]He is a Postdoctoral Research Associate in the
- [00:01:09.630]Departments of Animal Science
- [00:01:11.310]and Biological Systems Engineering here at UNL.
- [00:01:14.760]He uses precision rangeland management
- [00:01:17.850]to inform the application of adaptive grazing strategies.
- [00:01:21.180]And that is what he will be telling us about today,
- [00:01:23.250]so that's very exciting.
- [00:01:25.080]He was previously focused on remote sensing applications
- [00:01:27.480]in agriculture as well, such as cropland
- [00:01:29.910]and rangeland monitoring and management.
- [00:01:33.450]And then he is a fairly recent graduate.
- [00:01:36.180]In December of 2023, he graduated here from UNL
- [00:01:39.466]with a PhD in Natural Resources.
- [00:01:42.720]And he was also affiliated with
- [00:01:44.730]the Department of Biological Systems Engineering
- [00:01:46.740]at that time.
- [00:01:48.030]He is already published in peer-reviewed journal articles,
- [00:01:50.448]mostly in this area.
- [00:01:53.040]And so I, without further ado, thank you so much
- [00:01:56.340]for joining us today
- [00:01:57.420]and we look forward to your presentation.
- [00:01:59.760]Thank you.
- [00:02:00.720]Thank you Chris for the introduction
- [00:02:03.240]and thanks everyone for joining my presentation today.
- [00:02:06.930]Yeah, I'm Biquan.
- [00:02:09.432]I'm a Postdoc in the Department of Animal Science
- [00:02:12.540]and Biological System Engineering.
- [00:02:14.649]And today I bring you a talk about the works
- [00:02:19.230]in my dissertation, which aims to help adaptive grazing
- [00:02:26.520]in Nebraska by using different rangeland monitoring
- [00:02:30.510]and assessing techniques.
- [00:02:32.850]So at the...
- [00:02:43.080]At the beginning of this presentation,
- [00:02:46.260]I will express my gratitude to the people,
- [00:02:50.550]grant, and organizations list in this slide.
- [00:02:54.630]Thanks for their support, patience, and feedback
- [00:02:58.710]during my PhD.
- [00:03:00.300]And without their contribution,
- [00:03:02.340]you may be listening to another people speaking here.
- [00:03:06.660]And anyway, it's my time now,
- [00:03:10.140]so let's get started
- [00:03:12.480]and I will give you a big picture of the background
- [00:03:15.810]for my work.
- [00:03:18.060]'Cause I'm talking about adaptive grazing,
- [00:03:21.000]so I have to quickly go over some facts
- [00:03:24.060]about beef production in Nebraska.
- [00:03:27.390]Nebraska is the top three beef state in the US
- [00:03:31.740]and cattle industry can approximately $12 billion
- [00:03:36.990]every year to Nebraska economy.
- [00:03:40.260]And there are 6.5 million heads of cattle
- [00:03:43.680]and calves in Nebraska,
- [00:03:46.320]supported by the plant production from 22 million acres
- [00:03:51.330]of rangelands and pasturelands.
- [00:03:54.960]Grazing in rangelands and pastureland forms
- [00:03:58.890]a grazing system.
- [00:04:00.960]And in a grazing system there are components,
- [00:04:04.470]including climate and weather, grazing animals,
- [00:04:08.430]forage, management, soils and other components.
- [00:04:13.290]These components interact mutually, which can influence
- [00:04:17.790]the benefit of the grazing system
- [00:04:21.240]and the health of larger grassland ecosystem.
- [00:04:25.830]In a conventional grazing system,
- [00:04:28.650]a pasture is grazed by livestock for the entire season
- [00:04:33.450]without a recovery period, as shown here,
- [00:04:37.980]from mid-May to October.
- [00:04:41.370]This kind of pasture are large
- [00:04:44.280]and have relatively few water source for livestock,
- [00:04:48.780]which often result in concentrate grazing
- [00:04:53.100]with some low use efficiency of forage production
- [00:04:56.880]for the entire pasture.
- [00:04:59.010]And it can cause overgrazing, grassland degradation,
- [00:05:03.570]and other type of ecological issues.
- [00:05:07.350]And this is the reasons
- [00:05:09.330]that we developed adaptive grazing system.
- [00:05:13.380]These systems are decided to operate grazing
- [00:05:17.103]adaptively in accordance with forage supplied
- [00:05:22.140]by rotating livestock grazing over time and space,
- [00:05:26.850]across different oceans of the pasture.
- [00:05:31.200]And for example, in a deferred rotation
- [00:05:36.270]grazing system, here, the enlarged pasture is divided
- [00:05:41.670]into several smaller paddocks
- [00:05:44.760]and the cattles are moved over time and space
- [00:05:48.090]among this smaller paddock.
- [00:05:50.940]By doing this, we can maximize the use efficiency
- [00:05:55.590]of forage production and help the sustainability
- [00:06:01.310]of the grassland ecosystem.
- [00:06:04.800]And however, we are facing two important questions
- [00:06:09.885]when running the adaptive grazing system,
- [00:06:13.470]which are when to graze and where to graze.
- [00:06:17.790]So my work is trying to provide multiple source information
- [00:06:23.370]to help answer these two questions
- [00:06:26.263]for better adaptive grazing in Nebraska
- [00:06:31.920]by monitoring and assessing range conditions
- [00:06:35.790]using different techniques.
- [00:06:38.040]And in my dissertation,
- [00:06:41.250]there are three parallel project
- [00:06:44.430]using this different techniques.
- [00:06:47.730]The first project aims to improve our understanding
- [00:06:52.440]in the patterns and trends in forage production for years.
- [00:06:56.217]And in the second project, we estimate the in-season
- [00:07:01.380]forage production over time and space.
- [00:07:04.980]The goal of the third project is to gain insight
- [00:07:09.450]for the cattle grazing behavior and distribution.
- [00:07:14.430]And in the last 30 minutes, I will give you
- [00:07:19.260]more details about these three projects.
- [00:07:22.440]And for the first project,
- [00:07:24.150]we implemented a long-term exclosure monitoring
- [00:07:28.530]in Sandhills for improved understanding
- [00:07:32.970]of the patterns and trends in forage production for years.
- [00:07:39.750]In adaptive management,
- [00:07:41.700]learning from monitoring is a critical step.
- [00:07:45.540]And in the context of adaptive grazing,
- [00:07:49.170]how can grazing be adaptive without lowering the dynamic
- [00:07:54.390]of the rangeland, such as hand production
- [00:08:01.350]and communities, and other important matches?
- [00:08:05.160]And long-term monitoring comes to play
- [00:08:08.220]because it allows us to continuously collect the information
- [00:08:13.770]from the rangeland.
- [00:08:19.170]And as we note,
- [00:08:20.640]plant production is in variations over years.
- [00:08:25.950]In the Great Plains region,
- [00:08:28.050]many long-term grassland monitoring projects
- [00:08:31.200]have been initiated since 2000
- [00:08:34.350]because of the climate projection of rising temperature
- [00:08:39.060]and the more frequent chance in land use and land power.
- [00:08:43.920]From the literature,
- [00:08:45.600]we found that the total plant production
- [00:08:50.160]in the Great Plains region can vary 40% annually
- [00:08:55.770]as a response to the variability in precipitation
- [00:09:01.170]and temperature.
- [00:09:04.230]The variation in total plant production is
- [00:09:08.400]a collective response of plant community
- [00:09:12.930]and plant functional groups to the weather conditions.
- [00:09:19.560]And here is the figures showing the varying response
- [00:09:24.240]between C3 grasses and C4 grasses
- [00:09:27.810]to the annual precipitation and annual temperature.
- [00:09:31.800]In addition, plant production is influenced
- [00:09:36.000]not only by the weather condition within a season of a year,
- [00:09:41.850]but also by the across season weather condition.
- [00:09:46.770]For example, forbs were found to be more productive
- [00:09:52.440]in the following year after a drought year.
- [00:09:56.580]So there is some reasons
- [00:10:00.900]that why we are monitoring the Sandhill.
- [00:10:05.190]Firstly, I have mentioned that
- [00:10:09.120]there are 22 million acres of rangelands
- [00:10:13.320]and pasturelands in Nebraska.
- [00:10:15.990]And nearly 50% of them are located in Sandhill.
- [00:10:22.380]And secondly, the mid-grass prairie of Sandhill provide
- [00:10:27.660]the diverse habitat for 720 plant species,
- [00:10:32.910]including C4 grasses, C3 grasses, forbs, shrubs,
- [00:10:38.070]sedges, and other groups.
- [00:10:40.230]So we require detailed information in plant production
- [00:10:45.360]for years and for various plant functional groups
- [00:10:49.470]to help us for the decision making
- [00:10:54.660]for adaptive grazing strategy.
- [00:10:59.490]The objective of this study is to enhance our understanding
- [00:11:04.140]of the long-term chance and response
- [00:11:06.600]for plant functional group,
- [00:11:08.760]plant production in the Sandhill.
- [00:11:11.670]So by using the 15-year exclosure monitoring dataset,
- [00:11:17.730]we explore the trends for plant production,
- [00:11:21.930]including the total plant production
- [00:11:25.230]and the production for different plant functional groups.
- [00:11:29.790]After that, we investigate a response
- [00:11:33.240]of plant production to the resting season
- [00:11:36.360]and across season weather variables.
- [00:11:42.150]This long-term monitoring project was conducted
- [00:11:46.950]at Gudmundsen Sandhills Laboratory
- [00:11:51.960]in the central Sandhill.
- [00:11:54.000]And the date long-term monitoring showed
- [00:11:57.990]in this presentation is from 2007 to 2021.
- [00:12:03.120]We took the biomass measurement twice a year
- [00:12:07.020]in mid-June and mid-August.
- [00:12:10.260]And in addition, here is a weather station
- [00:12:15.360]at Gudmundsen provides us the daily precipitation
- [00:12:19.500]and temperature data for the analyze.
- [00:12:24.450]We harvest the plant material from 30 plants is cultured
- [00:12:30.630]every year and the harvest plant materials was separated
- [00:12:35.790]into six plant functional groups,
- [00:12:38.820]including C4 grasses, C3 grasses, forbs, shrubs,
- [00:12:44.700]sedges, and native annual grasses.
- [00:12:48.180]And by doing so, we...
- [00:12:50.790]And this is a 2D view of the exclosure.
- [00:12:55.907]So we harvested the plant material one side
- [00:13:00.180]of the exclosure in mid-June and another side in mid-August.
- [00:13:05.430]By doing this, we have June harvest biomass
- [00:13:09.510]for early season growing period
- [00:13:12.540]and we have August harvest biomass
- [00:13:16.320]for a full season growing period.
- [00:13:19.500]And then we can calculate the biomass change
- [00:13:22.950]for a late season growing period
- [00:13:25.860]from mid-June to mid-August.
- [00:13:29.640]And here are some weather variables used in this project.
- [00:13:34.140]We calculate the precipitation accumulation
- [00:13:37.800]and growing degree day accumulation, respectively,
- [00:13:41.910]as an individual weather variable.
- [00:13:45.180]And we, additionally,
- [00:13:46.950]calculate the standardized precipitation
- [00:13:49.680]and repair transpiration index, SPIE,
- [00:13:53.520]as a combined weather variable
- [00:13:56.400]because the calculation of SPIE use
- [00:14:00.300]both precipitation and temperature data.
- [00:14:04.860]And this slide shows all the weather variable
- [00:14:09.150]used in this study.
- [00:14:10.710]You may notice there is an index of zero, one, two,
- [00:14:15.823]for the weather variable.
- [00:14:18.210]And this index is used to specify
- [00:14:22.950]the weather conditions within and across season.
- [00:14:28.050]For example, an index of zero indicates
- [00:14:32.430]the weather condition of the current growing season
- [00:14:36.690]and an index of one indicates the weather condition
- [00:14:40.813]of growing season one year prior,
- [00:14:44.040]which means the previous year.
- [00:14:46.290]And index of two indicates the weather condition
- [00:14:50.310]of growing season two year prior.
- [00:14:54.058]And in this project we used the MK test
- [00:14:58.290]for the trend statistic.
- [00:15:00.480]After that, we built multiple linear regressions with the...
- [00:15:06.270]We built multiple linear regressions model
- [00:15:09.150]with the best subset method.
- [00:15:12.060]And we used the adjusted R-squared and BIC
- [00:15:16.530]to determine the best model for the response
- [00:15:20.280]of plant production to the weather condition.
- [00:15:24.900]And here are the results.
- [00:15:26.910]This table shows the trends in weather,
- [00:15:31.707]and precipitation and temperature.
- [00:15:34.410]And we observed a significant increasing trend
- [00:15:38.970]in growing degree days accumulation
- [00:15:42.720]during the early season period,
- [00:15:45.690]which indicates an early season warming trend in temperature
- [00:15:50.460]during the 15-year study period.
- [00:15:54.750]And however, we observed low significant chance
- [00:15:59.370]in the precipitation.
- [00:16:01.320]In addition, here is a weak increasing chance
- [00:16:05.040]in temperature for the full season,
- [00:16:08.310]but we may need more data points
- [00:16:11.640]to validate this trend.
- [00:16:16.350]So it indicates the importance of keeping
- [00:16:22.260]long-term monitoring in the Sandhill region.
- [00:16:26.730]And this graph here shows the chance in plant production.
- [00:16:31.738]The total grassland production is in variations
- [00:16:37.800]over years and over seasons with a mean variation
- [00:16:42.720]of 20% annually throughout the entire 15 years period.
- [00:16:50.100]And from this graph, we observe a substantial reduction
- [00:16:56.940]of plant production in C4 grasses in 2012.
- [00:17:02.610]However, in 2013, here is a remarkable increase
- [00:17:08.130]in the forbs production.
- [00:17:11.610]And this table shows more details
- [00:17:14.190]about the trends in the plant production.
- [00:17:18.780]We observed a significant decreasing trend
- [00:17:24.240]in the C4 grass plant production
- [00:17:28.200]both in mid-June and mid-August.
- [00:17:31.410]And an additional significant trend was observed
- [00:17:35.970]in shrubs with an increasing trend.
- [00:17:41.970]Here this box plot shows the trends in biomass percentage
- [00:17:48.150]of plant functional groups to the total production.
- [00:17:52.560]Here we've found that C3 grasses, C4 grasses, and forbs
- [00:17:58.710]together contribute nearly 90%
- [00:18:02.251]of the total grassland plant production.
- [00:18:05.100]These results align well with our current knowledge
- [00:18:09.450]of the characteristic of mid-grass carried in the Sandhill.
- [00:18:16.867]And in addition, we observe an abnormal increase
- [00:18:22.986]for the forb biomass percentage, up to fivefold in 2013.
- [00:18:32.730]And since the C3 grass, C4 grass, and forbs are
- [00:18:37.287]the three main plant functional groups in the Sandhill,
- [00:18:40.890]so we show the detailed trends for their biomass percentage
- [00:18:46.470]in this table.
- [00:18:47.850]And surprisingly we found that the biomass percentage
- [00:18:52.560]of C4 grass is decreasing both in mid-June and mid-August.
- [00:18:59.400]And for the C3 grasses biomass percentage,
- [00:19:02.880]it inhibit an increasing trends during the 15-year period.
- [00:19:13.050]And from the next few slides,
- [00:19:16.590]I will show you the model of the response
- [00:19:19.290]of the three main plant functional groups,
- [00:19:23.220]C4 grasses, C3 grasses, and forbs.
- [00:19:26.790]And for the C4 grasses, the meaningful models here
- [00:19:32.340]in late season period shows that C4 grasses is different
- [00:19:38.730]by the late summer, is different by the weather condition
- [00:19:45.360]of late summer of current year.
- [00:19:48.240]And particularly a wet late summer based on
- [00:19:53.190]the coefficient of the weather variable in this table.
- [00:19:59.295]And for the C3 grasses, we found that it's
- [00:20:04.022]preferred wet growing season of one year prior
- [00:20:10.590]and the early warming of the current year.
- [00:20:15.240]And for the forbs,
- [00:20:20.940]dry growing season one year prior possibly
- [00:20:25.740]drive the third response in the Sandhill region.
- [00:20:32.100]So here is the conclusion for this project.
- [00:20:37.200]Based by using a long-term exclosures monitoring dataset,
- [00:20:42.750]it provide us a detailed plant production information
- [00:20:48.030]in the trends of plant production
- [00:20:51.210]and its response to the weather for 2007 to 2021,
- [00:20:56.940]which contributes an overview for us to make the strategies
- [00:21:03.390]for adapted grazing in Nebraska.
- [00:21:09.990]And in the second project,
- [00:21:13.230]we used in season drone remote sensing
- [00:21:16.440]to estimate forage production over time and space.
- [00:21:22.560]In adaptive grazing, here is an important step,
- [00:21:29.370]modified management when needed.
- [00:21:34.050]But the question is how can we know when
- [00:21:39.450]the modification of management is needed?
- [00:21:43.620]So we need the truth of in-season evaluation
- [00:21:48.630]and this start this project was conducted
- [00:21:52.830]in the cool-grass period in eastern Nebraska,
- [00:21:57.930]where the dominant species is smooth bromegrass
- [00:22:02.790]introduced C3 or cool-season grasses.
- [00:22:09.270]Smooth bromegrass is used as forage
- [00:22:13.050]not only in eastern Nebraska
- [00:22:16.170]but also in some areas in state of Kansas
- [00:22:20.580]and South Dakota.
- [00:22:22.530]Being used as a forage, smooth bromegrass has
- [00:22:26.730]some advantage.
- [00:22:28.890]For example, it's best suited for early grazing
- [00:22:33.600]because it matures in late spring
- [00:22:37.620]and primary because of its late maturity in late spring,
- [00:22:42.650]it has relatively higher forage quality compared
- [00:22:47.370]to other C3 grasses.
- [00:22:49.980]In addition, smooth bromegrass is palatable to livestock
- [00:22:54.960]with an average digest benefit of 67%.
- [00:23:01.650]However, using smooth brome as forage has a disadvantage.
- [00:23:10.613]For example, it result in a short grazing window.
- [00:23:17.580]For example, in eastern Nebraska,
- [00:23:21.810]a typical grazing season spans from May to October
- [00:23:27.750]and this is the simulated production distribution curve
- [00:23:33.300]for smooth bromegrass.
- [00:23:35.370]As you can see here, there are two grazing window.
- [00:23:40.920]One is from May to June
- [00:23:43.350]and another is from September to October.
- [00:23:47.520]And these two grazing window together
- [00:23:51.120]only cover 50% period of the grazing season.
- [00:23:56.190]So we have applied management, such as fertilization
- [00:24:01.884]and rotating grazing and others
- [00:24:05.700]in the smooth brome dominate pasture.
- [00:24:10.020]But fundamentally, to maximize the forage utilization
- [00:24:18.150]of smooth brome's production,
- [00:24:21.180]we need an in-season forage evaluation
- [00:24:25.110]or a timely forage production estimation
- [00:24:28.860]so that we can make the adaptive grazing strategies
- [00:24:34.770]according to the production of smooth brome.
- [00:24:41.700]And there are some approach that we have used
- [00:24:45.090]for forage production measurement on the ground level.
- [00:24:49.530]for example, by using rising plate meter
- [00:24:52.800]or equipping the plant material
- [00:24:55.620]to within a sampling quadrant.
- [00:24:57.990]However, although this mass, this approach give us
- [00:25:04.590]a reliable forage production data,
- [00:25:09.120]it may not appropriate for frequent in-season evaluation
- [00:25:15.300]because those approach are plant- and labor-consuming.
- [00:25:20.250]And we can own, and as you note, ranch in Nebraska can up
- [00:25:25.800]to hundreds or thousands of acres.
- [00:25:29.040]And we can only sample a small size of data
- [00:25:38.520]across the entire pasture.
- [00:25:41.160]And based on the dispersed samples,
- [00:25:44.610]we may not be able to represent the true forage conditions
- [00:25:50.760]for the entire pasture
- [00:25:52.590]because of the lack of spatial details.
- [00:25:57.720]But drone remote sensing could be a potential tool
- [00:26:02.550]for us to help the in-season rangeland monitoring.
- [00:26:08.820]For example, in agriculture in cropland management,
- [00:26:13.740]drones have been used to remotely sense images
- [00:26:17.970]from a canopy power, from a canopy level
- [00:26:22.110]over the growing season,
- [00:26:26.310]and then with estimation algorithms,
- [00:26:29.640]there are some estimated biomass maps produced
- [00:26:34.350]so that we can use for cropland management,
- [00:26:38.070]including spraying fertilizers or for disease
- [00:26:43.350]and pest control.
- [00:26:45.420]So it's a potential to use the drone remote sensing
- [00:26:49.530]for the in-season rangeland monitoring
- [00:26:52.680]to help us to provide more useful data information
- [00:26:58.170]for us to answer the question of when to graze
- [00:27:02.430]and where to graze in a pasture.
- [00:27:06.870]So the objective of this study is to establish the use
- [00:27:11.400]of drone remote sensing for in-season forage
- [00:27:15.090]biomass estimation.
- [00:27:17.220]We evaluate the estimation performance
- [00:27:20.550]for bromegrass biomass with drone image
- [00:27:24.660]by using a random forest regression model.
- [00:27:28.230]And then we assessed the chance in the biomass production
- [00:27:33.690]from the drone estimate maps over time and space.
- [00:27:38.370]This study, this project was conducted
- [00:27:41.550]at Eastern Nebraska Research Extension and Education Center.
- [00:27:47.460]The study area is 15 acres,
- [00:27:50.610]encompassing two sites
- [00:27:54.090]and in each size there are three management treatments,
- [00:27:58.680]including controlled treatment, supplemental treatment,
- [00:28:03.330]and fertilization treatment.
- [00:28:05.640]So this treatment is expected
- [00:28:08.160]to make the biomass production,
- [00:28:14.700]make the variation in biomass production over time.
- [00:28:19.574]And these three management pastures were further fenced
- [00:28:27.030]into smaller paddocks for rotational grazing.
- [00:28:33.630]And of course, the 15 acres study site we used
- [00:28:38.460]the quadrant to sample the biomass production
- [00:28:42.840]in late May, late June, and late July in 2022
- [00:28:50.160]and a total of 106 samples were sampled.
- [00:28:55.200]And in each samples we have that biomass
- [00:28:58.860]for brown vegetation and we have live biomass
- [00:29:03.000]for green vegetation as well as the total biomass
- [00:29:07.560]by adding that biomass and live biomass together.
- [00:29:14.730]Here this slide shows the drone data collection.
- [00:29:19.770]So we flew a drone at 100 meters above ground level.
- [00:29:26.910]This drone carrier RGB and the multi-spectral camera
- [00:29:31.830]from the RGB camera, which obtains the digital surface
- [00:29:36.810]model which represents the forage height.
- [00:29:40.950]And from the multi-spectral image,
- [00:29:44.790]we derived the vegetation in this.
- [00:29:50.910]By using tent predictor variables in this project,
- [00:29:55.260]we built random forest regression model
- [00:29:58.920]to estimate the three biomass in this study.
- [00:30:03.690]And a total of 106 samples were randomly split
- [00:30:09.210]into a training set and a testing set.
- [00:30:13.410]In the testing set we used the R-squared, IMSE,
- [00:30:17.847]and rRMSE for the model evaluation.
- [00:30:23.010]Here are the results.
- [00:30:25.410]These three scattering plot shows the
- [00:30:29.790]estimation performance in the testing set.
- [00:30:33.780]So we found that the R-squared for that biomass was 0.63
- [00:30:40.350]and for live biomass is 0.85
- [00:30:44.610]and for total biomass is 0.81.
- [00:30:48.300]So these results shows that the estimation
- [00:30:51.900]of live biomass outperformed another two biomass.
- [00:30:58.590]And here is the estimated biomass mapped over time
- [00:31:04.440]and over space from the drone remotely-sensed image.
- [00:31:10.500]For that biomass, we observe an increasing trend
- [00:31:15.450]over time from late May, late June, and late July.
- [00:31:21.300]This is expected.
- [00:31:22.890]And for the live biomass and total biomass,
- [00:31:27.330]it inhibits decreasing trends over time.
- [00:31:32.100]And in spatial, we found that
- [00:31:36.990]there is high biomass production in the treatment
- [00:31:43.680]of fertilization treatment.
- [00:31:46.860]And there is a median production
- [00:31:49.050]in the supplemental treatment with a low production estimate
- [00:31:54.330]in the controlled treatment.
- [00:31:56.250]So that result makes sense.
- [00:32:00.300]And so when go to the next step
- [00:32:04.890]to look at the perform,
- [00:32:07.140]to compare the performance of the brome methods
- [00:32:11.190]and the drone estimate methods.
- [00:32:13.890]So because the estimation performance
- [00:32:18.180]in live biomass outperform,
- [00:32:20.850]so I used the live biomass data here
- [00:32:26.370]in this box plot to show they're different
- [00:32:31.080]in the two approach.
- [00:32:33.840]The box plots above is used the ground samples
- [00:32:38.670]from quadrant.
- [00:32:40.380]And the box plot here at the bottom is based
- [00:32:44.790]on the pixel from the estimate
- [00:32:47.790]from the drone estimate maps.
- [00:32:49.980]So we observe consistent trends and patterns
- [00:32:53.730]from these two box plots.
- [00:32:55.890]And over space and in a spatial scale,
- [00:33:03.618]there is a high production in fertilization treatment
- [00:33:08.610]and median production in supplemental treatment
- [00:33:11.760]and low production in controlled treatment.
- [00:33:15.450]So based on our results,
- [00:33:19.350]we found that the results showed that drones can be
- [00:33:23.370]a useful tool to facilitate a fast and efficient
- [00:33:28.410]in-season estimation of the smooth bromegrass
- [00:33:32.641]storage biomass.
- [00:33:34.530]And from the drone estimate map, biomass map,
- [00:33:39.600]it shows temporal and spatial biomass variation,
- [00:33:44.190]which provides useful information for us
- [00:33:47.850]to answers the question of when to graze and where to graze.
- [00:33:55.620]And then in the third project,
- [00:33:58.950]we use the GPS collar to check the cattle grazing behavior
- [00:34:05.010]and distribution.
- [00:34:08.640]In a grazing system, grazing animals is
- [00:34:12.480]the most essential component.
- [00:34:15.600]And thinking from the cattle perspective,
- [00:34:22.620]thinking from the cattle point of view
- [00:34:25.410]would be beneficial for us
- [00:34:27.420]to develop the adaptive grazing strategy.
- [00:34:34.350]So here in a rotational grazing system,
- [00:34:38.280]the cattles are moved over time and space
- [00:34:42.523]across different, smaller paddocks.
- [00:34:46.380]So a question is are there any changing behavior
- [00:34:51.600]when they are rotated over time and space,
- [00:34:55.140]when they are facing frequent changing environment?
- [00:35:00.720]And this, the answer to these questions will help us
- [00:35:06.030]to specify the adaptive grazing strategy
- [00:35:12.360]more better according to the cattle demands in the pasture.
- [00:35:20.460]And actually, animal checking is not a new topic.
- [00:35:26.370]And long time ago in history,
- [00:35:30.270]producers and scientists used visual inspections
- [00:35:38.100]in the pasture or in the range to check the cattle movement.
- [00:35:42.570]And in 1980s, the radio text or high frequency device used
- [00:35:49.470]more frequently for the animal research.
- [00:35:53.970]And over the past 20 years,
- [00:35:56.460]GPS collar have been widely used in animal science research,
- [00:36:02.760]such as wildlife habitat research
- [00:36:06.090]and something like that.
- [00:36:09.000]So the objective of this study is
- [00:36:12.870]to investigate the trends in grazing behavior
- [00:36:17.934]of cattle during a deferred rotational system.
- [00:36:22.770]And it supports the factors that would result
- [00:36:26.040]in these trends.
- [00:36:28.350]In specific, we use a branch of GPS checking data
- [00:36:34.440]to assess the daily travel distance of cattle
- [00:36:38.400]during the rotation in early period, mid period,
- [00:36:45.960]and late period.
- [00:36:47.580]And then we investigate the potential factors
- [00:36:51.120]and effect by developing
- [00:36:53.280]a resource selection functions model.
- [00:36:58.260]We call it four herds of cattle
- [00:37:02.790]at the Barta Brothers Ranch in the Sandhill
- [00:37:06.390]from May to September in 2016 and 2017.
- [00:37:13.283]In each herd, they are rotated.
- [00:37:19.050]They were rotated in a particular pattern
- [00:37:23.068]during a specific period.
- [00:37:26.550]For example, from the Herd A,
- [00:37:30.090]they first rotated in N3 paddocks
- [00:37:33.120]and then moved to the N3 paddocks for the mid rotation,
- [00:37:41.070]and then eventually they moved to the N1 paddock.
- [00:37:46.110]And in this project, we used the Lotek GPS collar
- [00:37:51.300]to collar three to five heads of cattles per herd.
- [00:37:56.550]And these collar record the position location
- [00:38:00.570]of the cattle at the interval of every 10 minutes.
- [00:38:06.780]The GPS collar stayed lock this kind of data
- [00:38:12.360]and there is a algorithm for us
- [00:38:15.906]to identify the cattle behavior of grazing and resting.
- [00:38:25.110]And in this project, we only use the grazing data point
- [00:38:29.670]for the analyze because we are checking
- [00:38:32.400]their grazing behavior.
- [00:38:34.830]And to answer the question, are there any change
- [00:38:39.090]in grazing behavior during the rotation?
- [00:38:42.990]We analyzed the daily travel distance.
- [00:38:50.310]After that, we established the resource selection function
- [00:38:56.490]model to investigate the factors
- [00:39:00.600]for the grazing behavior.
- [00:39:03.420]Resource selection function is the largest regression
- [00:39:07.230]with the binary response variable.
- [00:39:10.230]And in the equation here,
- [00:39:13.260]the predictor variables are the resource
- [00:39:16.110]or the factors associated with the cattle grazing behavior.
- [00:39:22.740]And with the coefficient, we can calculate the odds ratios.
- [00:39:30.660]And odd ratios greater than one means
- [00:39:35.069]there are more chance to be grazed
- [00:39:37.110]of a location by the cattle.
- [00:39:40.710]And this might list the five potential factors
- [00:39:45.600]investigate in this study.
- [00:39:48.000]So we have NDVI as a percent of the forage condition
- [00:39:53.310]and we have distance to water, distance to fence,
- [00:39:57.270]and we have topography wetness index,
- [00:40:01.080]and topography position index.
- [00:40:03.780]And because these two index are highly correlated.
- [00:40:08.700]So in the result in the next few slides,
- [00:40:12.480]we only keep one of this index here as the factors.
- [00:40:19.530]So this graph shows the daily travel distance.
- [00:40:26.520]And the answer is yes,
- [00:40:28.890]there are significant change in daily travel distance
- [00:40:32.640]during the location, which means the cattles
- [00:40:35.640]are changing their behavior during the rotation.
- [00:40:39.840]And we found that in the earlier rotational period,
- [00:40:45.963]they traveled the most up to 4000 meters daily.
- [00:40:56.100]And but in the mid period they decreased
- [00:40:59.790]their travel distance to 3500 meters.
- [00:41:05.340]And in the late period it declined further.
- [00:41:08.670]So one potential reason for this is when they moved
- [00:41:14.820]to the later pattern in mid period and late period,
- [00:41:20.260]there are a more availability of forage production.
- [00:41:26.730]so that they don't have to move, travel a lot
- [00:41:32.700]to search for the forage.
- [00:41:37.440]And in this figure also shows there is
- [00:41:40.860]no significant difference among herds
- [00:41:43.980]and no significant between the data of 2016 and 2017,
- [00:41:51.457]which indicates that the herd effect
- [00:41:56.700]and the annual effect won't influence our results
- [00:42:02.340]in this project.
- [00:42:05.310]And when we look at the odds ratio for the factors here,
- [00:42:10.530]this thoughts show that in the early grazing period
- [00:42:15.630]in 2017, the NDVI have an extremely high odds ratio
- [00:42:24.450]up to 27.
- [00:42:26.700]And this means the NDVI or the condition
- [00:42:32.100]of forage production significantly drive
- [00:42:36.510]the cattle grazing behavior during this time.
- [00:42:40.770]And in 2017 is quite a dry season.
- [00:42:47.580]So potential reason here is the cattle move
- [00:42:53.040]more in this period for search of fresh
- [00:42:58.350]or green vegetation for their eating.
- [00:43:06.600]And this is extremely high value,
- [00:43:10.350]make this odds ratio cast together.
- [00:43:14.430]And here is a zoomed in beautiful odds ratio.
- [00:43:21.000]And similarly we found that in 2016 NDVI
- [00:43:28.020]in early grazing season is pretty,
- [00:43:31.230]has a pretty high odds ratio as well,
- [00:43:35.460]greater than one.
- [00:43:37.020]So this result indicates that
- [00:43:40.560]in the early grazing season,
- [00:43:42.720]NDVI is the most important factors difference
- [00:43:47.100]the cattle grazing behavior.
- [00:43:50.250]And then when we look at the June end yield here,
- [00:43:54.429]we found that in the mid-grazing period
- [00:43:57.750]and the late-grazing period distance to water
- [00:44:01.350]and distance to fence is the most important,
- [00:44:06.300]the most significant factors.
- [00:44:10.020]And I mean in these two periods there are pretty
- [00:44:16.290]much availability of forage production
- [00:44:20.040]and the temperature already increased during this period.
- [00:44:24.120]So the cattle would want to stay more
- [00:44:30.810]around the water source.
- [00:44:33.300]So this means this result in this in our project.
- [00:44:42.150]So as a conclusion here,
- [00:44:44.760]by using the GPS checking data,
- [00:44:47.790]we gained the insights of the trends
- [00:44:52.980]in cattle grazing behavior over time
- [00:44:55.890]and space in a deferred rotational grazing system.
- [00:45:00.480]And this information can help us better
- [00:45:03.960]to understand the cattle demands during rotation
- [00:45:11.640]so that we can make the adaptive grazing strategy
- [00:45:16.260]in accordance with their demands.
- [00:45:20.340]So here as a discussion how have
- [00:45:25.680]the street techniques contributed to the adaptive grazing
- [00:45:30.048]in my dissertation?
- [00:45:32.190]So by using a long-term exposure monitoring,
- [00:45:35.850]we improved, we enhanced our understanding of the patterns
- [00:45:40.740]and trends in forage production for years.
- [00:45:43.890]It provides an overview for us to develop
- [00:45:48.810]the adaptive grazing strategy.
- [00:45:52.080]And in the second project,
- [00:45:54.810]we used the in-season drone remote sensing
- [00:45:58.355]to estimate the forage production
- [00:46:02.280]and assess the variation over time and space
- [00:46:07.830]so it can contribute some useful information
- [00:46:12.120]for us to answer the question when to graze
- [00:46:15.270]and where to graze.
- [00:46:16.920]And in the third project, we used the GPS collar
- [00:46:21.360]to check the cattle grazing behavior
- [00:46:24.930]and distribution in a rotational grazing system.
- [00:46:30.600]And so in the future, we should continue
- [00:46:35.190]the long-term monitoring and try
- [00:46:37.950]to integrate the joint remote sensing
- [00:46:40.950]during the grazing season and growing season.
- [00:46:44.370]And then we should keep using the GPS collar
- [00:46:49.170]for tracking the cattle-grazing behavior.
- [00:46:52.980]And how about the more futures?
- [00:46:56.190]How can we develop a system
- [00:46:58.830]with these techniques built in?
- [00:47:01.680]Because these three projects are parallel
- [00:47:05.520]and they are conducted in different location,
- [00:47:09.240]in different system.
- [00:47:11.100]So if we can make this technique built in
- [00:47:14.580]and develop another adaptive grazing system,
- [00:47:20.070]it will more useful for us to better
- [00:47:25.620]for precise rangeland management.
- [00:47:29.010]And I can foresee, there are many potential
- [00:47:32.580]to use this technique for adaptive grazing.
- [00:47:36.270]And yeah, thank you for your listening
- [00:47:40.950]and I think I kind of went out of time.
- [00:47:44.575]Yeah, yeah.
- [00:47:46.565]Thank you.
- [00:47:53.087]Thank you, Biquan.
- [00:47:53.970]Any questions?
- [00:47:56.280]Do we have any questions online, Norma?
- [00:47:57.903]Okay.
- [00:48:05.070]Yeah, thanks for the talk.
- [00:48:08.280]It's nice to see a grazing talk in here.
- [00:48:12.300]Yeah, so I got a couple questions
- [00:48:13.920]about the behavioral stuff.
- [00:48:15.300]And I don't remember if I asked you, this in your defense.
- [00:48:17.760]Yes. But yeah,
- [00:48:19.860]so hopefully I'll ask you some
- [00:48:20.970]different questions this time.
- [00:48:22.977]So I guess my biggest question is about
- [00:48:28.745]why you used resource selection function?
- [00:48:36.060]So, but before I get to that, I guess
- [00:48:38.520]how did you measure daily movement
- [00:48:42.270]if you used resource selection, which is point pattern
- [00:48:46.350]analysis, how did you do distance?
- [00:48:50.760]So actually in, as for your question,
- [00:48:54.360]so actually in this project we don't use the resource
- [00:48:58.470]selection function for the delete movement.
- [00:49:01.860]We just statistic, we do a statistically analyze
- [00:49:06.390]for the daily travel distance.
- [00:49:10.980]Okay.
- [00:49:11.813]Yeah, I got that.
- [00:49:12.646]So I guess my question is why not just
- [00:49:15.090]use path analysis for all of it?
- [00:49:18.120]'Cause you're doing paths here, right?
- [00:49:19.822]Yeah.
- [00:49:21.865]Okay, so then what was the point of resource selection,
- [00:49:24.780]'cause the two are different.
- [00:49:27.240]So I guess I'm just curious like why
- [00:49:29.670]the difference between the two
- [00:49:31.560]'cause they have different underlying assumptions?
- [00:49:34.590]Yeah, so we do the travel distance
- [00:49:39.570]to answer the questions.
- [00:49:41.820]Are there any changing behavior during the rotation?
- [00:49:47.610]So the Daily Travel Distance is a
- [00:49:51.160]more clear matches to show their change
- [00:49:57.030]over time and over space.
- [00:50:01.337]And then with the random forward regressions,
- [00:50:03.510]we want to answer the questions of what are
- [00:50:07.710]the factors different, the change.
- [00:50:13.463]Cool, so I guess I'll ask you a more direct question.
- [00:50:15.990]So how do you know that that's actually grazing
- [00:50:19.260]and not just movement?
- [00:50:21.540]Yeah, this is a good question.
- [00:50:23.370]So here, is a developed algorithms for the this callers.
- [00:50:32.370]And this algorithms can distinguish the cattle
- [00:50:39.000]activity as grazing and resting.
- [00:50:43.980]So we only use the grazing data points
- [00:50:50.900]for our entire analyze.
- [00:50:53.414]So all the data points are based
- [00:50:54.690]on their grazing movement, the grazing behavior.
- [00:50:59.580]Is it clear?
- [00:51:03.540]Without a Z axis,
- [00:51:04.860]I don't know how you know
- [00:51:05.760]that they're actually grazing and not just moving
- [00:51:08.430]because I mean like their heads are down, right?
- [00:51:10.590]Yeah. They're grazing
- [00:51:11.997]and versus movement.
- [00:51:13.320]And I also don't really know like
- [00:51:14.790]without measures of tortuosity, like how does their...
- [00:51:19.110]Like they're not just moving in a straight line
- [00:51:21.210]for 4000 meters.
- [00:51:22.440]So without sensuosity measurements
- [00:51:25.440]or like whether they're head tilt is down versus up,
- [00:51:29.040]I don't know how you're distinguishing
- [00:51:30.870]between grazing and movement.
- [00:51:33.240]Yeah. Because to me
- [00:51:34.200]like grazing, the distance there is just movement,
- [00:51:37.320]it's not really behavior?
- [00:51:39.630]And so that's why I was asking
- [00:51:42.583]about resource selection functions and all this other stuff.
- [00:51:43.500]That's great.
- [00:51:44.940]But that's just an, we're just talking about movement.
- [00:51:47.670]We're not really talking about behavior.
- [00:51:50.700]So I'm may misunderstanding to showing this data here.
- [00:51:55.410]So actually, we lock more data by using this GPS code.
- [00:52:01.860]And there is another sensor to lock their grazing, yeah.
- [00:52:07.230]The accelerate meters to lock their grazing behavior.
- [00:52:13.740]And the algorithm is based on those added information
- [00:52:18.570]to distinguish their grazing activity and resting
- [00:52:24.300]or other activity.
- [00:52:26.790]Yeah, I hope this help a lot.
- [00:52:37.396]That was a very nice seminar.
- [00:52:39.900]I have a question about the quality
- [00:52:42.330]of the residue or the forage.
- [00:52:45.390]If the cattle have an option of eating forage
- [00:52:49.740]from an unfertilized or normal pasture
- [00:52:52.950]and one that's been fertilized, where do they go and why?
- [00:52:58.050]And yeah, this is a great interesting question.
- [00:53:02.700]So, and the answer to,
- [00:53:06.510]I don't have an exact answer for this questions.
- [00:53:09.990]But I think we can export more by using
- [00:53:14.100]some in-season monitoring techniques such as drone
- [00:53:18.840]to help answer this question
- [00:53:21.180]and it will provide us more specific
- [00:53:26.220]answer to the two questions that I missed
- [00:53:30.383]that I want to solve in my dissertation.
- [00:53:35.610]It would seem that if you have these GPS tags
- [00:53:38.910]on the animals, you could fertilize part
- [00:53:41.880]and not part and see where they go.
- [00:53:44.340]Yeah, yeah. And what they like.
- [00:53:46.290]Yeah, and I think this is,
- [00:53:48.300]that's why I'm saying there is more potential,
- [00:53:54.240]there have more potential to make these techniques
- [00:53:58.680]or technology together to make them build
- [00:54:03.780]in a adaptive grazing system.
- [00:54:06.270]And this would be,
- [00:54:07.530]I imagine this would be the direction
- [00:54:12.450]of the adaptive grazing development in the future.
- [00:54:19.650]I'm gonna answer Jim's question here
- [00:54:21.690]so you don't have to do the research to try
- [00:54:23.760]and determine the answer of it.
- [00:54:26.010]When the plants are vegetative,
- [00:54:27.240]they will definitely go to the fertilized area.
- [00:54:30.935]When the plants start to become very reproductive
- [00:54:34.560]and start forming stiff stems,
- [00:54:37.860]you'll see a lot more selection.
- [00:54:40.650]And if there's a big difference between the unfertilized
- [00:54:43.620]and the fertilized in terms of the STEM production,
- [00:54:46.620]they'll more often go to the unfertilized area.
- [00:54:52.500]So I have another question then, maybe for Bruce.
- [00:54:56.160]If you wander into a pasture,
- [00:54:58.020]you'll see cow pies out there
- [00:55:00.090]and probably urine deposits and this sort of thing.
- [00:55:03.330]Some really green grass around it,
- [00:55:05.190]but the cattle don't seem to like some of that.
- [00:55:07.620]What's going on?
- [00:55:11.280]You have an answer? I think you may.
- [00:55:19.350]I'll give you an interesting part of that is
- [00:55:22.410]that if those cow pies and urine spots were made
- [00:55:26.580]by a different species of animal,
- [00:55:29.220]they won't reject it nearly as much
- [00:55:33.780]as they do by their own species.
- [00:55:35.880]But it is basically an older type of thing
- [00:55:40.530]in terms of their rejection.
- [00:55:43.410]The cow pies are rejected much more aggressively
- [00:55:47.640]than the urine spots.
- [00:55:48.870]And in fact they'll, graze the urine areas pretty regularly,
- [00:55:55.860]especially if it's been like a season away
- [00:55:59.310]a few months away from when it was actually deposited.
- [00:56:10.260]Great talk, first of all.
- [00:56:12.000]But I was wondering in Random Forest,
- [00:56:14.670]why did you choose Random Forest
- [00:56:16.680]and what were your predictor variables
- [00:56:18.900]and which one explained the most variation in grazing?
- [00:56:24.060]Yeah, thanks for your question.
- [00:56:26.490]So I choose random forest regression
- [00:56:30.060]because I think the grazing behavior is kind of a,
- [00:56:36.180]a binary response and it will graze for long graze.
- [00:56:41.670]And this function can give us a likely hope
- [00:56:46.800]for their activities which makes more sense for their,
- [00:56:52.920]to quantify their grazing preference.
- [00:56:57.720]And then in my random,
- [00:57:01.440]in my resource selection functions
- [00:57:04.020]and we have the data of grazing points
- [00:57:10.710]and we have low grazing point as the critical variables.
- [00:57:18.090]And then the grazing point was a size of value of one,
- [00:57:22.870]but the learned grazing point is a size of value of zero.
- [00:57:28.050]So by using this, we can calculate a function such like that
- [00:57:32.476]and then the estimate, the result for there
- [00:57:38.190]for the output would be a value from zero to one.
- [00:57:43.830]So by using this, we can quantify the grazing preference.
- [00:57:48.940]And here are the five critical variables for this project.
- [00:57:58.620]We have NDVI as the forage condition
- [00:58:02.700]and we have distance to water and distance to fence
- [00:58:06.870]and other two variables, yeah.
- [00:58:21.240]You didn't do anything with soil moisture.
- [00:58:24.694]And particularly with early-season grazing,
- [00:58:29.460]that soil moisture at the beginning of the season would be
- [00:58:32.580]much more predictive than trying to use
- [00:58:38.400]the early season rainfall.
- [00:58:43.244]Any thoughts along that line
- [00:58:44.456]or ways that you might be able to overcome that limitation?
- [00:58:50.326]Yeah, this is great suggestions.
- [00:58:53.430]So actually here, the topography wetness index
- [00:58:59.130]is kind of some similar index for the soil moisture.
- [00:59:04.680]Because the topography wetness index,
- [00:59:08.460]it calculates the water accumulated in a location.
- [00:59:16.440]So I think it's kind of an indicator for the soil moisture.
- [00:59:23.400]And actually, in our models, in our model,
- [00:59:28.276]it has some...
- [00:59:38.850]Oh here, we just use the topography position index
- [00:59:43.058]because they are highly correlated
- [00:59:50.004]with the wetness index.
- [00:59:53.070]So I think probably I can try
- [00:59:57.510]to use some image of soil moisture
- [01:00:03.000]and then see the correlation again and run the model again
- [01:00:07.290]and see which factors which work most better.
- [01:00:13.230]Yeah, I think this is good suggestions.
- [01:00:16.410]And I have been to talk and this,
- [01:00:23.580]a professor told me that sometimes the wind
- [01:00:27.330]would be an important factor as well
- [01:00:30.480]because the down wind and the up wind directions
- [01:00:33.960]will influence the cattle behavior.
- [01:00:38.250]So I think it would be worth it
- [01:00:40.170]to increase the wind direction
- [01:00:43.410]and something data like that into the model
- [01:00:47.400]to make it more useful in the future.
- [01:00:52.290]Did you do any correlations
- [01:00:53.910]between those four different methods of systems there?
- [01:00:59.130]Your distance to water, to fence, TPI, and NDVI?
- [01:01:04.020]Yeah, I did the correlation
- [01:01:07.440]and we just found that the topography wetness index
- [01:01:12.030]and topography position index are highly correlated.
- [01:01:16.470]So we just keep one here, yeah.
- [01:01:22.468]Okay, I think we're at time.
- [01:01:24.120]Any one or not?
- [01:01:25.380]Okay.
- [01:01:26.213]All right, so thank you so much, Biquan,
- [01:01:28.950]for joining us today and for this wonderful seminar
- [01:01:33.210]and all these good discussions that were stimulated.
- [01:01:36.480]Thank you all for joining us this week
- [01:01:39.030]and we look forward to seeing you again next week.
- [01:01:42.210]Thank you. Thank you.
- [01:01:44.772](class applauding)
The screen size you are trying to search captions on is too small!
You can always jump over to MediaHub and check it out there.
Log in to post comments
Embed
Copy the following code into your page
HTML
<div style="padding-top: 56.25%; overflow: hidden; position:relative; -webkit-box-flex: 1; flex-grow: 1;"> <iframe style="bottom: 0; left: 0; position: absolute; right: 0; top: 0; border: 0; height: 100%; width: 100%;" src="https://mediahub.unl.edu/media/22323?format=iframe&autoplay=0" title="Video Player: Toward Adaptive Grazing: Leveraging Different Rangeland Monitoring And Assessing Techniques" allowfullscreen ></iframe> </div>
Comments
0 Comments